hasaneyldrm/exercises-dataset

The **Exercises Dataset** repository by hasaneyldrm is a comprehensive, production-ready fitness exercise dataset containing 1,324 unique exercises, each paired with an animated GIF, a 180×180 thumbnail image, and rich structured metadata including category, body part, equipment type, target muscle, synergist muscle groups, and step-by-step instructions available in 10 languages: English, Spanish, Italian, Turkish, Russian, Chinese, Hindi, Polish, Korean, and French. The dataset is stored as a JSON array in `data/exercises.json`, formally validated by a JSON Schema (Draft 2020-12) file, and all media assets are sourced from Gym Visual under specific attribution terms alongside an MIT license for the code and data. The repository is notably the data backbone powering **LogPress**, an AI-assisted workout tracking mobile application, demonstrating real-world production use rather than being a purely academic or experimental resource. Coverage across body parts is broad, with upper arms (292), upper legs (227), back (203), waist (169), and chest (163) being the most represented categories, and equipment ranges from body weight (325 exercises, ~25% of the total) to dumbbells, cables, barbells, leverage machines, bands, Smith machines, kettlebells, and more. Two standalone HTML tools are bundled with the repository: `index.html`, a fully client-side exercise browser with live search, category/equipment/target filtering, infinite scroll, and multilingual instruction display; and `setup.html`, a developer setup guide that generates ready-to-run SQL INSERT statements for SQL Server, PostgreSQL, MySQL, and SQLite, provides copy-paste API client code in JavaScript, Python, C#, Java, PHP, Go, and cURL, and includes structured LLM prompts for generating complete REST APIs using frameworks like Express.js, FastAPI, ASP.NET Core, Spring Boot, Laravel, and Gin. The data schema is thorough, with each record carrying a unique ID, full exercise name, multilingual instruction strings and step arrays, primary and secondary muscle fields, equipment, media paths, attribution, and an ISO 8601 creation timestamp. The project is notably developer-friendly, requiring no server to browse or explore, and is well-suited for fitness app development, exercise recommendation machine learning models, health research, and educational prototyping. Its multilingual scope, visual media assets, formal schema validation, and integrated developer tooling make it a notably feature-rich and immediately deployable dataset in the fitness and health technology space, distinguishing it from simpler exercise lists or single-language alternatives.